To construct a new calibration method that combines usability and accuracy for estimating herbage mass from rising‐plate meter readings, we derived four models differing in the way their parameters are related to sampling date and compared their estimation accuracies using cross‐validation. The parameters of the linear regression for each sampling date showed seasonal variations, which had a steep decrease from early April to early June and a gradual increase thereafter. The pooled models were less accurate for estimating herbage mass than a separate model, which had specific parameters for each sampling date (S model). Among the pooled models, however, those in which the parameters were assumed to be linear functions (PL model) or combined functions (PC model) of the sampling date showed substantively improved estimation accuracy compared with the traditional pooled model, in which the parameters were assumed to be fixed throughout the year (PF model). Moreover, at the beginning of the season, the models derived from previous years' data were suggested to be applicable as a practical method. Thus, it can be concluded that these types of pooled calibration could be used as ‘compromise methods’ that combine both accuracy and usability.

Although the precision of herbaceous biomass estimation depends on the sample number, the spatial heterogeneity of the biomass, and sampling procedures, the magnitudes of the influences on the precision have not been clarified. We simulated virtual plant communities based on the gamma distribution to clarify the relationships between the precision of estimating herbaceous biomass and the number of samples, sampling density, spatial heterogeneity of the biomass, and sampling procedures. Using only two parameters, the gamma distribution can approximate the frequency distribution of herbage mass with varying heterogeneity. Our simulations demonstrated that the number of samples is a more influential factor than sampling density on the precision of the herbaceous biomass estimation. Moreover, our simulations confirmed that biomass heterogeneity strongly affected the precision and quantified the magnitude of the influence. When we estimated biomass with random sampling and a 50 3 50 cm quadrat and accepted estimation error of 6 10% of the mean for a confidence interval of 95%, the numbers of samples needed were 200, 77, and 9 for very, intermediate, and less heterogeneous grasslands, respectively. Similarly, when we estimated biomass with a ranked set sampling (RSS), then 24, 15, and 4 samples were needed in very, intermediate and less heterogeneous grasslands, respectively. We came to two conclusions: 1) In less heterogeneous grasslands, good precision of estimation can be obtained with a small number of samples, and it is useful to employ RSS. The cutting method, as well as nondestructive methods, will be practical; and 2) estimation for heterogeneous grassland requires a large number of samples, and it is not so useful to employ RSS. For that reason, more research is needed on nondestructive methods.

To evaluate the carbon (C) sequestration function of grassland soils in Japan, soil C stocks were measured in 24 grasslands (3–43‐year‐old pastures) across 14 livestock farms nationwide. Soil C stocks varied among soil types, and the values in the upper 25 and 50 cm were higher in Andosols (mean, 12.4 and 19.3 kg m−2, respectively) than in Brown Forest soils (7.5 and 13.7 kg m−2) and other soil types (5.5 and 7.5 kg m−2). At the same time, C stocks varied among pastures within each soil type. Compared to data from the published work on the C content shortly after pasture establishment, aged pastures had decreasing C concentrations as the soil depth increased, suggesting substantial C accumulation in the top soil layers during pasture aging. This C accumulation caused grassland soils to store as much C as adjacent forest soils. Although the C stocks in the grassland soils were not statistically different from those in the adjacent native forest soils, some grassland areas stored greater amounts of C than the forests, indicating a possibility of increasing soil C stocks through improved grassland management.

In order to clarify the effect of land‐use change from forest to grazing pasture on the organic carbon storage in Andosol soil, the Rothamsted carbon turnover model for volcanic soil was applied to a pasture situated at the National Livestock Breeding Center (37°09′N, 140°03′E). The top 25‐cm soil layer was considered to be an active soil carbon pool. The carbon storage in the soils of native forest surrounding the pastures ranged 111–163 t C ha−1 with an average of 133 t C ha−1, which was adjusted according to an equivalent soil weight of pasture. The pasture soil carbon stocks ranged 88–135 t C ha−1, with variations according to site and/or pasture age. The carbon inputs to the soil through the above‐ and below‐ground dead material from pasture plants and cattle feces were estimated to be 1.1, 1.8 and 0.9 t C ha−1 year−1, respectively. As the model outputs of 14C content of the soil, which is an index of carbon dating corresponding to nuclear weapons testing, showed a relatively close agreement with the observations, the modeling was acceptable for the purpose of predicting the turnover of organic carbon in Andosol soil. The model simulation demonstrated that, in order to maintain the average forest carbon level, 3–4 t ha−1 year−1 of the organic carbon input would be needed. These inputs would be provided in a grazing pasture producing 8–9 t ha−1 year−1 of above‐ground dry matter.

We propose the following method for estimating above‐ground plant biomass (biomass, hereafter) in grassland as a variant of Shiyomi's (1991) second method and test its accuracy of estimation. The method proceeds thus: (i) place a given number of quadrats in the grassland of interest; (ii) an electronic capacitance probe is used to provide corrected meter reading (CMR) values within each quadrat; (iii) two sampling points with a biomass of C1 g and C2 g based on CMR values are selected to allow subdivision into three approximately equal frequency classes; (iv) quadrats are placed and CMR values are measured at the two sampling points; (v) the vegetation within the two quadrats is harvested and these samples weighed after drying; (vi) each quadrat (placed in step [i]) is divided into three categories based on the CMR values – less than C1 g, more than C1 g and less than C2 g, and more than C2 g; and (vii) biomass is estimated from the data obtained by these procedures using the gamma model. Test results suggest that the accuracy of biomass estimation of our method is equivalent to that of the cutting method using several quadrats. Thus, we propose the use of an electronic capacitance probe to rank pasture biomass and the selection of only two quadrat sites for cutting as this offers labor‐saving advantages over current methods.

This study aimed to estimate green herbage biomass (GBM) and crude protein (CP) concentrations of a mixed‐sown pasture in Hokkaido, Japan using ground based hyperspectral measurements and geostatistical analysis. The mixed‐sown pasture consisted of a relatively flat section renovated by over‐seeding a grass (Subunit 1, 2.6 ha) and a hilly aged section (Subunit 2, 5.0 ha). Hyperspectral reflectance and plant data were collected for 22 days in August 2009 from 88 plots within the two subunits. For mapping, separate spectral readings, without plant sampling, were obtained from a total of 347 plots along permanent transects in the pasture. Genetic algorithm‐based wavebands selection with partial least squares (GA‐PLS) regression analyses was performed to predict GBM and CP concentrations using both reflectance and first derivative reflectance (FDR) datasets. Then, geostatistical analysis with semivariograms was conducted to determine sampling interval of GBM and CP concentration in Subunits 1 and 2. In the GA‐PLS analysis, the most accurate results were obtained by calibration of GBM using FDR (cross‐validated coefficient of determination, ; cross‐validated root mean square error, RMSECV = 42) and of the CP concentration using raw reflectance (, RMSECV = 2.02). Geostatistical analysis with semivariograms showed that, at the landscape scale, the GBM patch sizes in Subunits 1 and 2 were 31 and 67 m, respectively, and those of the CP concentration were 37 and 54 m. These values indicate that the spatial distribution patterns of these pasture parameters were more heterogeneous in Subunit 1 than in Subunit 2. The analysis result also indicates that the sampling interval for GBM and CP concentration should be <15 m. This work shows the ability to estimate the spatial distribution of GBM and CP concentration for the implementation of site‐specific grazing management.

A modified method with visual observation for estimating biomass distribution on grasslands is proposed. This labor‐saving technique facilitates surveys for estimating herbaceous biomass distribution for grasslands. It is based on the principle of Shiyomi's visual observation method. The procedure is performed as follows. (i) Set two points with biomass of c1 g and c2 g used as criteria in a pasture (c1 < c2). (ii) In the first visual observation, divide herbaceous biomass in the pasture into two classes that are more or less than c1 g. (iii) In the second visual observation, divide herbaceous biomass in the same pasture into two classes that are more or less than c2 g. Then, match the trial numbers obtained in the first and second observations. (iv) Measure the biomass weights of c1 and c2 with cutting. (v) From the data obtained above, infer the herbaceous biomass distribution using the gamma model. The procedure was conducted in a Zoysia grazed pasture. The following are discussed: advantages and regulations of the current method with a gamma model; some problems of the cutting method, as viewed from the shape of herbaceous biomass distribution; and the influence of grazing pressure on herbaceous biomass distribution.

A rising plate meter (RPM), which enables rapid, non‐destructive estimation of herbage mass of pastures, is useful in grazing management. Herbage mass is estimated from the RPM readings using a linear regression model, which has the parameters α (constant) and β (slope). In this study, we analyzed the seasonal changes in the two parameter values for total and green herbage (TH and GH, respectively) masses in Zoysia‐dominated pastures, and examined the potential for incorporating a periodic function (PF) of the parameters into the equation for estimating herbage mass. The values of both α and β showed a weak seasonal periodicity for TH. No periodicity and a clear seasonal periodicity were detected in the α and β values for GH, respectively. Compared with the linear regression models developed separately for individual dates, the equation incorporating a PF for β improved the estimation accuracy for the GH mass (from 0.64 to 0.77 as R2), though it did not affect the predictability for the TH mass. Incorporation of a PF into the model relating herbage mass to the RPM reading increases the sample number and broadens the data range by allowing the use of all seasonal data, and produces a single equation useful for estimating herbage mass across seasons.